The present invention relates to nuclear boiling water reactors (BWRs) that utilize natural circulation and more particularly to enabling load-following capability in such reactor designs.
Existing large BWRs are of the forced-circulation type. In BWRs undergoing power generation operations, reactor coolant, initially in the form of sub-cooled liquid (e.g. water), is circulated by main coolant recirculation devices (e.g. jet pumps or mixed-flow motor-driven pumps) around a path a portion of which is comprised of the core lower plenum region (located at the bottommost section of the reactor), thence through the nuclear core and into a core upper plenum in communication with the core. Flow exiting the core upper plenum then passes through standpipes that lead to an assembly of steam separators. The reactor coolant exiting the nuclear core and passing into the core upper plenum is a two-phase mixture of steam and water, the proportion of which varies depending upon such factors as the power output from the fuel bundles, the amount of sub-cooling present in the coolant entering the fuel bundle, and the amount of flow through the bundles. This last factor depends on the power of the recirculation pumps and the hydrodynamic flow resistance presented by the fuel bundle geometry and wetted surfaces, and the amount of orificing representing restrictions to flow just prior to the coolant's entrance into the core fuel assembly.
Joining with the core effluent in the core upper plenum is the core "by-pass" flow, which is reactor coolant that has flowed from the core lower (entrance) plenum into the region external to the fuel assembly channels (but inside the core shroud), thence upwardly generally through the region occupied by cruciform-shaped control blades which stand in various degrees of insertion into the core, thence across the upper grid member (called the "top guide") which with its lattice-like configuration keeps the fuel assemblies in a regular array, and finally into the core upper plenum. This by-pass coolant stream at its discharge into the core upper plenum is comprised substantially of saturated liquid, with perhaps a small amount of steam. Within the core upper plenum, these two effluents--by-pass flow and fuel bundle exit flow--rapidly mix together and quickly lose identity from their origins.
Mechanical steam separation can be utilized to accomplish the separation of the steam from the steam/water mixture exiting the core. Some earlier BWR designs used free-surface steam separation where, just as in the household tea kettle, steam separates unaided from the free-surface, and saturated water remains in the bulk coolant, which in BWRs is recirculated back down the downcomer annulus. This type of steam separation is feasible so long as the steam-leaving velocity, i.e. the bulk average velocity of the steam taken across the available pathway flow area, is not large, i.e. is no greater than about 1.8 foot/second. If steam-leaving velocities exceed this value, there tends to be carried along with the steam an unacceptably high moisture content. The high moisture levels saturate the moisture-drying abilities of the steam dryer, thus resulting in an unacceptably high moisture content in the steam leaving the reactor and supplied to the turbine. When steam moisture contents are too high in the turbine steam flow, accelerated erosion can occur on first-stage turbine blades and the efficiency of the turbine is reduced.
It is possible to obtain free-surface separation capabilities if the reactor pressure vessel (RPV) cross-sectional area is made sufficiently large. However, cost economies dictate that minimum diameter RPVs be used, so that mechanical steam separation has been developed to handle the high power output steam production levels of modern BWRs. In these latter designs, the steam bulk average velocity moving through the wet steam plenum region immediately downstream of the mechanical steam separators is about 5 feet/second.
The fuel assemblies grouped over the central region of the core tend to have higher exit steam qualities than do bundles located at the peripheral region of the core. It is desirable, nonetheless, that the flow rates and steam/water mixture proportions entering the steam separator standpipes be relatively uniform. To facilitate gaining more nearly uniform steam/water mixture for entry into the standpipes, the standpipe entrances are separated from the fuel assemblies by a distance of, for example, about 5 feet. Turbulent mixing occurring between the plumes leaving adjacent fuel assemblies, each with a different void content, is one mechanism acting to produce a more nearly uniform mixture which enters into the steam separator standpipes. More important to achieving flow mixture uniformity, however, is the hydrodynamic flow resistance represented by the standpipes, each with their end-mounted steam separators. Complete flow mixture uniformity entering the standpipes is at best difficult to achieve and, even with a five-foot separation between fuel assembly exits and standpipe entrances, it is not a design basis used for reactor performance evaluations.
The steam separator assembly consists of a domed or flathead base on top of which is welded an array of standpipes with a three-stage steam separator, for example, located at the top of each standpipe. One function of the standpipes is to provide a stand-off separation of the larger-diameter steam separators, which are generally arranged in a particularly tightly-compacted arrangement in which external diameters of adjacent separators are nearly touching with each other, so that separated liquid coolant discharged at the bottom of the separator has a more "open" flow path outwardly from the reactor longitudinal axis and out to the downcomer annulus region which lies at the inboard periphery to the RPV. A second purpose for the standpipes in a high-power-output natural-circulation reactor using mechanical steam separators is to provide juxtaposed regions which promote natural-circulation by means of a vertical region of two-phase (and, thus, low-density) coolant inside the standpipes which is juxtaposed against single-phase liquid coolant outside the standpipes in a so-called "downcomer region", in which region height provides a very significant part of the total natural circulation driving head for coolant flow circulation within the reactor.
The steam separator assembly rests on the top flange of the core shroud and forms the cover of the core discharge plenum ("core upper plenum") region. The seal between the separator assembly and core shroud flange is a metal-to-metal contact and does not require a gasket or other replacement sealing devices. The fixed axial flow type steam separators have no moving parts and are made of stainless steel, for example, to resist corrosion and erosion.
In each separator, the steam/water mixture rising through the standpipes (the "standpipe region") impinges upon vanes which give the mixture a spin, thus establishing a vortex wherein the centrifugal forces separate the water from the steam in each of three stages. Steam leaves the separator at the top of this assembly and passes into the wet steam plenum below the dryer. The separated water exits from the lower end of each stage of the separator and enters the pool (the "downcomer region") that surrounds the standpipes to join the downcomer flow. The steam exiting from all separators either may be in the same horizontal plane, or the separators may be arranged in a slightly crowned fashion at the center to compensate for a crowned water gradient of the pool surrounding the standpipes.
The steam separator assembly may be bolted to the core shroud flange by long hold-down bolts, or the separator together with the dryer assembly may be held down onto the core shroud flange by contact from the reactor head when the latter is assembled to the reactor vessel. The nominal volumetric envelope of the steam separator assembly is defined by the horizontal plane of its lower flange that contacts the core shroud flange, its cylindrical sides that provide part of the five-foot stand-off from the fuel assembly exits, the circumscribed diameter of the outermost row of standpipes, the circumscribed diameter of the outermost row of steam separators, and the generally horizontal plane of the exits to the steam separators.
The core upper plenum region in a BWR currently under design known as the "simplified boiling water reactor" (SBWR) is substantially devoid of other mechanical devices, pipes, or structures; whereas the core upper plenum of a BWR/6 and "advanced boiling water reactor" (ABWR) reactor design generally contains spargers and nozzles for core sprays, and distribution headers for core flooders, respectively. In both reactor types, these spargers/headers are located at the outer periphery of the core upper plenum, mounted below the core shroud flange so that the sparger/header is clear of the refueling removal path of peripheral fuel assemblies and, thus, do not become removed during core refueling operations.
With specific reference to a natural-circulation SBWR, it will be observed that there are no recirculation pumps to aid in coolant recirculation. Steam generation in the core produces a mixture of steam and water which, because of steam voids, is less dense than saturated or sub-cooled water. Thus, the boiling action in the core results in buoyancy forces which induce core coolant to rise upwardly, to be continuously replaced by non-voided coolant arriving from beneath the core in the core lower plenum region. As the coolant leaves the core, it rises through the core upper plenum region, then through the standpipes region, and finally into the steam separators. This voided mixture inside these standpipes continues to be less dense than non-voided coolant external to the standpipes, resulting in the development of additional buoyancy force to further drive the coolant circulation. That this process is quite effective in promoting coolant recirculation can be noted from reported tests made in forced-circulation power reactors where the coolant circulation pumps are shut off. Even with their relatively short steam separator standpipes, reactor power levels of 25% and coolant flow rates of 35% of rated flow, are readily and safely maintainable.
The SBWR reactor is but modestly different from the forced-circulation BWR, with the most prominent differences being that the standpipes region is to be considerably longer in the SBWR (to develop a higher differential head), the core overall height may be somewhat shorter (for example, being 8 or 9 feet active fuel length versus 12.5 feet active fuel length in recent forced-circulation reactors), and the core power density will be somewhat lower. The severity of orificing--a means to promote hydrodynamic stability--at the entrance to the BWR fuel bundles may be lessened. The fuel bundle may have a larger diameter fuel rod in, for example, a 6.times.6 rod array, whereas the rod array for a forced-circulation reactor often is an 8.times.8 rod array. The design flow rates per fuel bundle, and the flow rates per steam separator, will be somewhat reduced in the SBWR design. Fuel exit steam quality will be approximately the same between the two designs. In the SBWR reactor design, no spargers or discharge headers are installed in the core upper plenum, while in the ABWR reactor, spargers or discharge headers are installed in the upper core plenum.
In some versions of SBWR reactors under study, the standpipes are very long while the core upper plenum is short. In other versions, the converse is true. The present invention is applicable equally in either version.
"Load-following" is the action of bringing the power output of a BWR into balance with an incrementally changed power output demand. This demand change arises from the electrical grid to which the nuclear power station is coupled and represents a change from prior steady-state (balanced) operating conditions.
By way of illustration, assume that an SBWR is operating at 90% of rated power output. Existing within the core will be some certain distribution of voids, i.e., steam vapor in the form of steam bubbles. The lowermost parts of the fuel assemblies will contain non-voided coolant because of the sub-cooled liquid conditions existing in the core lower plenum, the source for water entering the core. Partway up the flow path within the fuel assemblies, steam generation begins, so that a steam/liquid mixture develops with the steam proportion rising with increasing travel upward through the fuel assembly. Control blades immediately outside the fuel assembly channels will stand in various degrees of withdrawal from the core depending on the particular point the core has reached in its fuel cycle lifetime.
The steam output from the nuclear boiler is coupled to a turbine generator, which, in turn, is coupled electrically to the grid. A nuclear boiler pressure regulation control system is installed, the action of which changes the position of turbine steam control valves in such a way so as to maintain constant the nuclear boiler pressure as measured in the reactor steam dome.
A change in grid electrical demand--say an incrementally increased demand for more electrical power from the power station--causes a signal to be sent to a control rod positioning system that results in incremental withdrawal of certain of the control blades still not fully withdrawn from the core. This withdrawal has the effect of making the reactor temporarily more reactive, allowing an increase in neutron flux, that, in turn, produces a higher rate of nuclear fission throughout the fuel rods. The thermal capacitance represented by the mass of the fuel material (uranium dioxide) briefly, i.e. for a few seconds, absorbs the thermal energy produced throughout the fuel rods as their internal temperature rises. (The fuel heat transfer thus lags the neutron flux, the transient response characteristic being that of, typically, a seven-second time constant.) Soon, however, the higher temperatures lead to greater heat transfer from the now-higher fuel clad temperature to the reactor coolant, and so an incrementally higher amount of steam is formed. In addition, the point where boiling first begins within the fuel assembly moves slightly downward in response to the higher heat transfer that is occurring. This combination of incrementally more voids in prior boiling regions, plus downward movement of the boiling boundary, now introduces negative reactivity effects that returns the reactor to a balanced, steady-state, power level, but one that is generating incrementally more steam. In response to the larger steam generation rate, to hold pressure constant in the reactor steam dome (as the control system mandates), the pressure regulation system progressively incrementally opens the turbine control valves, thus releasing a net greater quantity of steam to the turbine. Higher steam rates passing through the turbine produce the required incremental increase in ultimate response to the initiating electrical grid demand for more electrical power from the station.
The principles described above can be extended to those conversant with nuclear engineering practice to understand other types of power adjustments. It will be apparent that the foregoing illustration is the response to a small increase in load demand. Obviously, the adjustment of reactor power output also can be performed manually by the reactor operator, through his actions to cause control blades to be inserted farther into the core or withdrawn farther outward from the core.
Often, a nuclear power station is required to sustain larger load demand adjustments than the relatively small adjustment described above. Existing nuclear power stations are deficient in that it takes time to retract the control blades. Even when the control blades are moved in groups ("ganged rod movement"), it still requires time for the groups to be sequentially moved. An additional disadvantage to load-following by control blade movement can be that the heating transients within the fuel occurring close to the ends of where the control blades are positioned, over time can produce undersirable stress-cycling on fuel cladding.
An alternative load regulation means that has been found effective for forced-circulation BWRs is to use recirculation flow control. A signalled change in reactor power demand is sent to a control system that adjusts the recirculation flow upward or downward. The recirculation flow is regulated either by changing the speed of the main recirculation pumps, or in other applications by throttling the output from constant-speed pumps by means of a flow-control valve. The changed flow causes a rather prompt change in the amount of voids in the core and a similar change in the position of the boiling boundary within the fuel assemblies within the core. For example, the action response in the recirculation flow control system to an incremental demand for more reactor output would be to raise the rate of recirculation flow. This sweeps some existing voids out of the core, and raises the position of the boiling boundary. In turn, neutron flux rises, fission rate increases, and shortly a higher total amount of steam is being generated. With the reappearance of "near-normal" levels of voids in the core in response to the higher power output, the reactor condition returns to a "steady-state", but now at a higher output level. The two advantages of recirculation flow control are that the rates of change in reactor power can be faster; and since control blades are not required to be moved, no additional significant stress-cycling duty is imposed on the fuel rods.
To date, however, natural-circulation reactors have had only control rod movement available to them as a load-following means. As described earlier, a drawback with this mode of load-following control, i.e. performing load-following by moving control blades, is that it can be a slow-acting system because there are so many blades which must be moved some variable small amounts to effect a change while keeping the neutron flux profiles in desirable patterns. Thus, for various modes of power operation, it would be desirable for there to be other methods of more rapidly, yet controllably, affecting reactor power output, and thereby provide enhanced capability to perform a wider envelope of load-following maneuvers.